An On-Line Learning Method for Object-Locating Robots using Genetic Programming on Evolvable Hardware

Created by W.Langdon from gp-bibliography.bib Revision:1.7630

@InProceedings{seok:2000:arob,

• author = "Ho-Sik Seok and Kwang-Ju Lee and Byoung-Tak Zhang",
• title = "An On-Line Learning Method for Object-Locating Robots using Genetic Programming on Evolvable Hardware",
• booktitle = "Proceedings of the Fifth International Symposium on Artificial Life and Robotics",
• year = "2000",
• editor = "Masanori Sugisaka and Hiroshi Tanaka",
• volume = "1",
• pages = "321--324",
• address = "Oita, Japan",
• month = "26-28 " # jan,
• organisation = "International Society for Artificial Life and Robotics (ISAROB)",
• keywords = "genetic algorithms, genetic programming, FPGA",
• URL = "http://bi.snu.ac.kr/Publications/Conferences/International/AROB00.ps",
• URL = "http://citeseer.ist.psu.edu/456254.html",
• size = "4 pages",
• abstract = "Evolvable hardware is a new concept of FPGA which has a capability of dynamic reconfiguration during run time. Due to its dynamic reconfiguration ability, evolvable hardware can optimise itself through learning. In this paper, we present a method for learning robot controller on evolvable hardware. For learning, we employ genetic programming. Typically, genetic programming uses tree-structured representation. However, tree structures are inconvenient for crossover in hardware and tend to consume much resource. Therefore, we use a linear chromosomes to represent genetic trees on evolvable hardware. The learning objective of the robot is to locate a light source avoiding obstacles.",
• notes = "AROB-00",

}

Genetic Programming entries for Ho-Sik Seok Kwang-Ju Lee Byoung-Tak Zhang

Citations